Hippocampus

Examining the Human Hypocampal Formation

Buried deep within the medial temporal lobe of the human brain lies the hippocampal formation—a group of many millions of neurons organized into a network quite different from that found anywhere else in the nervous system. It is a striking structure whose bulb-like shape, protruding into the lateral ventricles, has has drawn the attention of anatomists since the first dissection that took place in Egypt.

The hippocampal formation is a group of brain areas consisting of the dentate gyrus, hippocampus, subiculum, presubiculum, parasubiculum, and entorhinal cortex. The basic layout of cells and fiber pathways of the hippocampal formation is much the same in all mammals (Fig. 1x).

1. One set of structures, often defined as the hippocampal region, consists of the CA fields of the hippocampus itself, the dentate gyrus, and the subiculum.
2. The second set of structures consists of the adjacent cortical structures that lie along the parahippocampal gyrus. This set consists of the entorhinal, perirhinal (occasionally divided into temporopolar and posterior regions), and parahippocampal cortices.

Along with the adjacent cortical structures in the parahippocampal gyrus, the human hippocampal region is critically involved in memory for facts and events (explicit or declarative memory).

Importance of the Hippocampal Formation

There are several reasons the hippocampus has captivated scientists across diverse fields—the hippocampus has something for everyone. Perhaps you are a psychologist interested in memory, a synaptic physiologist investigating neuronal and synaptic plasticity, or a computational neuroscientist wanting to build a neural network model, the hippocampus and its associated structures are an attractive set of brain structures on which to work.

In parallel, clinicians concerned with the basis of neurological conditions such as epilepsy or Alzheimer’s disease had their attention drawn to the hippocampal formation because of the pathological processes observed to occur there and the opportunities that scientific study of this area of the brain offers for novel therapeutics.

Over the last many years, the pyramid-shaped cells of the hippocampus have become the most intensively studied neurons in the brain. As a result, we know a great deal about their development, synaptogenesis, neurotransmitter receptors and ion channels, micro-circuitry, and cell biological machinery.

No less impressive has been the molecular analysis of neurotransmission in hippocampal cells; what has been learned from this has revealed general properties that seem to apply in other areas of the central nervous system. We also know something about why and when these cells are activated in the living brain.

Along the extensive dendritic arborizations of these pyramidal cells, there can be many thousands of miniature dendritic spines. These are the sites where most of the excitatory synapses are to be found. An important finding is that the efficiency with which these excitatory synapses transmit messages can vary as a function of neural activity. Many suspect that synaptic plasticity is a key mechanism of memory.

Role of the Hippocampus in Human Memory

The hippocampus is engaged in remembering information that can be described in a propositional or declarative manner. It would be almost impossible to begin a discussion of the role of the hippocampal region in human memory without considering the patient H.M.

Patient H.M.

The striking discovery, over many decades ago, that patient H.M. had a relatively pure memory deficit after surgical excision of the relief of epilepsy has had a profound effect on the study of memory and, through that, on our understanding of the functions of the hippocampus itself.

Although scattered hints in the literature prior to this suggested that damage to the hippocampal region might affect memory, the striking discovery that patient H.M. had a relatively pure memory deficit after surgical excision of the medial temporal lobe had a profound effect on the study of memory Opens in new window and, through that, on neuropsychologists understanding of the functions of the hippocampus itself.

Following a brief initial report (Scovlle, 1954) describing the profound memory loss that resulted from the surgery, Scoville and Milner (1957) presented a detailed neuropsychological assessment of H.M. and nine other patients with varying degrees and locations of resection. This seminal work concluded with the following text.

Subsequent testing of H.M. was able to reveal not only the breadth and severity of H.M.’s memory impairment but, critically, it also showed a vast array of cognitive and mnemonic function that was unaffected by the lesion of the temporal lobe.

The pattern of data allowed Milner (Milner et al., 1968; Milner, 1972) to make several conclusions that expanded on those noted above and helped lay the foundation for subsequent investigation into the amnesic syndrome. First, Milner noted that damage to the medial portions of the temporal lobe results in anterograde amnesia Opens in new window—inability to acquire long-term memory for new facts or events. Although memories from early life appear to be intact. Conversely, there was some clear loss of memory of information acquired for some time prior to the operation—a condition known as retrograde amnesia Opens in new window).

Together, these observations indicated that the medial temporal lobe might not be a permanent repository for memory but that it plays a time-limited, albeit critical, role in memory. Second, she noted that there was no loss in general intellect or perceptual ability, indicating clear dissociation between memory and other aspects of cognition.

Third, she noted that short-term memory remained intact, indicating clear dissociation between immediate, or “working,” memory and permanent, long-term memory. Finally, she noted that damage to the medial portions of the temporal lobe did not abolish all forms of long-term learning and memory, indicating that the medial temporal lobes were not required for at least some forms of long-term memory.

A large amount of the research on the human hippocampus has been aimed at functionally dissociating the role of the hippocampus from the role of adjacent cortical structures. A popular idea draws on the anatomy to suggest that the hippocampus integrates information from and combines the processing of the adjacent cortical structures that feed into the hippocampus.

Two fundamental hypotheses that share this basic idea and that are both driven by data from human and nonhuman studies have been proposed and explored.

One hypothesis states that there is a clear dissociation of function between the hippocampus and the adjacent structures. For example, the hippocampus has been described as being involved in memory that is associational, multi-items, spatial, episodic, and recollective, whereas the perirhinal cortex (and by extension at times the parahippocampal cortex) is involved in memory that is automatic, noneffortful, single-item, and familiarity, or recency-based (in contrast to recollective), with this distinction being qualitative rather than quantitative (Brown & Aggleton, 2001).

The other hypothesis states that the dissociation of function is more quantitative than qualitative in nature and that the hippocampus and the adjacent structures in the parahippocampal gyrus are all broadly involved in declarative memory. By virtue of being farther up in the hierarchical structure of the medial temporal lobe, the hippocampus is proposed to “combine and extend” the processing carried out by the entorhinal, perirhinal, and parahippocampal cortices (Squire & Zola-Morgan, 1991).

By virtue of receiving input from both perirhinal and parahippocampal cortices (via the entorhinal cortex), the hippocampus is in a position to be able to integrate information across these structures and sources of information. Thus, “associative,” or “conjunctive,” processing can be attributed to the hippocampus. However, the structures in the parahippocampal gyrus also receive input from a wide range of sources, putting them in a position to perform “associative” or “conjunctive” processing as well.

As the input to the hippocampus consists of more refined and further processed information (its major input arrives from the entorhinal cortex, which receives approximately two-thirds of its input from the perirhinal and parahippocampal cortices), the hippocampus is in a position to perform different, potentially more abstract or complex associative or conjunctive processing. This is not to say, however, that there is binary dissociation of function between the hippocampus and the adjacent cortical structures according to associative or conjunctive versus single-item, episodic versus semantic, recollection versus familiarity, and so on.